Using the Soil-Adjustment Vegetation Index from Landsat-8 Imagery for Estimating the Nutrient Content of Oil Palm Leaves for Optimized Fertilizer Application

¹ Agricultural Science Doctoral Program, Faculty of Agriculture, Universitas Sumatera Utara, Jl. Dr. A. Sofian No.3, Sumatera Utara, Indonesia
² Faculty of Agriculture, Universitas Sumatera Utara, Jl. Dr. A. Sofian No.3, Sumatera Utara, Indonesia
³ Sydney Institute of Agriculture, C81 - Biomedical Building, The University of Sydney, Australia

^* Corresponding author: t.sabrina@usu.ac.id

Abstract

Fertilization plays a vital role in oil palm cultivation, contributing 30–35% of total production costs. Rising fertilizer prices make precise nutrient management essential to maintain productivity and sustainability. Conventional fertilization practices are often generalized and do not consider spatial variability in plant nutrient status, leading to inefficiencies in dosage recommendations. This study aims to optimize oil palm fertilizer application by integrating remote sensing and geospatial analysis using Landsat-8 imagery. Three soil-adjusted vegetation indices SAVI, OSAVI, and MSAVI were applied to estimate the leaf nutrient content of Nitrogen (N), Phosphorus (P), Potassium (K), and Magnesium (Mg). The estimated nutrient content was used to determine fertilizer dosage based on the harvest-transported nutrient approach. The study was conducted in oil palm plantations in North Sumatra Province, Indonesia. Results showed that the MSAVI index had the highest correlation with nutrient content (R² = 0.75 for N), enabling the generation of site-specific fertilizer recommendations. The optimal fertilizer doses were 2.50–3.00 kg palm⁻¹ year⁻¹ for Urea, 1.75 kg palm⁻¹ year⁻¹ for TSP, 2.25–2.75 kg palm⁻¹ year⁻¹ for MOP, and 3.25–3.50 kg palm⁻¹ year⁻¹ for Dolomite. The findings demonstrate that integrating remote sensing with geospatial analysis provides an efficient and accurate approach for precision fertilization in oil palm management.